Skip to content
Preprint

Stop or Clarify? Experience-Conditioned Active Fact Verification

Oct 2026 · 0 citations · 32 references
Computer Science

TL;DR

Sequential Experience-Adaptive Clarification wraps a fixed verifier with a lightweight evidence-acquisition policy that improves mean accuracy and macro-F1 over the compared fact-verification and active-retrieval baselines and supports selective clarification as a contributor to the gain.

Abstract

Fact-verification benchmarks score the final label, while an interactive verifier must first decide whether the available evidence is sufficient. We focus on that intermediate Stop-or-Clarify Decision. Uncertainty alone is insufficient, because the benefit of another evidence-acquisition step is unknown before the step is taken. Sequential Experience-Adaptive Clarification (SEAC) wraps a fixed verifier with a lightweight evidence-acquisition policy. For each claim, SEAC retrieves evidence, runs the verifier, forms a label-free decision state, and combines stop risk with an observational clarification-utility estimate from earlier executed trajectories. Clarification is requested only when this score exceeds the action cost and decision threshold. On SciFact, LIAR, and FEVER, SEAC improves mean accuracy and macro-F1 over the compared fact-verification and active-retrieval baselines. Equal-budget and formula-level diagnostics support selective clarification as a contributor to the gain, while the small SciFact split leaves its paired improvement inconclusive.

View source

Similar papers

#artificial intelligence Review Oct 2026

ActiveEvidence: Decision-Focused Evidence Graphs with Verifiable Acquisition Gaps

A multimodal agent must decide whether visual evidence is worth acquiring, which candidate or resolution to query, and when further interaction can no longer change its decision. We study this problem under a strict free-metadata/paid-observation boundary, stochastic vector costs, and a frozen execution trace, and trea...

Miao-Bo Hu, Shu-Hao Hu, Xiao-Bo Guo et al. · 0 citations
Open access 2026

Um, Actually: Timely and Calibrated Fact-Checking with DIVER

Fact verification pipelines are often unrealistic: they assume full-input access, operate on isolated atomic claims, and overlook the central question of when a system should act under partial evidence. We introduce DIVER (Deliberative and Iterative fact VERification), an LLM -based framework for calibrated incremental...

Sen-Yi Zhang, Tasnim Kabir, Jordan L. Boyd-Graber · 0 citations
#artificial intelligence Review Sep 2026

Claim-Gated Source-Risk Auditing for Generative Search

The specification separates this endpoint from citation support and review priority, and binds decisions to versioned evidence spans, and defines the independent annotation, held-out evaluation, and paired utility tests still required to establish semantic validity and deployment benefit.

Kai-Nan Zhou, Chu-Hong Xu, Gang-Zhen Qian et al. · 0 citations

When Does Better Confidence Improve a Knowledge-Graph Audit?

Confidence estimates guide human review in LLM-built knowledge graphs, but an erroneous shared entity need not be the only reason an incident fact is wrong. We study how this distinction affects budgeted audits, alongside selection headroom and label validity. Modelling non-adaptive auditing as budgeted covering, we ch...

Zehao Lin, Kai Chen, Xi-Xuan Hao et al. · 0 citations
Open access Oct 2026

An Auditable Binary Evidence-Escalation Protocol for Prompt-Constrained Retrieval-Augmented Question Answering

Technical-support and engineering knowledge services must decide whether one selected passage is sufficient or additional retrieved evidence should enter the generation prompt. Aggregate accuracy and mean context size conceal whether expansion recovers omitted evidence, adds no value, or harms an otherwise correct answ...

Jin-Yi Zhu, Hao Wu, Yi Cao · 0 citations
Review Nov 2026

Auditing the Judge: Internal Signals for LLM Reasoning Assurance

Language-model systems use self-judgment to decide which answers to release, review, or recompute. Because intervention is costly, an assurance score must rank errors well at the available budget, not merely over the full ranking. We compare the judge's continuous verdict-token score with a supervised correctness probe...

Mayank Gulati, G. Wunder · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.